Having played a little bit with contours, I would say that that predefined ones may be ok for rough estimates but if we need to get detailed, there is just no way it will reflect the reality.
So here are a few questions :
1. Can I define a custom contour (script / macro ?) that I could then apply at will ?
Here's an example : I want to mobilize 10 dozers to open some access to my work site. For obvious reasons, I do not want to mobilize all 10 on Day 1 and I will have to stage their mobilization, say 2 at a time and every other day. So my 'ramp-up' will take 8 days and on the 9th day I have my 10 dozers working (it look like this : 2-2-4-4-6-6-8-8-10-10-...). Similarly, I plan to demobilize them 8 of then 2 by 2 but over just 4 days when the job is done and I will keep 2 for the remaining duration of my project, just in case I need to re-work some places (...-10-8-6-4-2-2-2-...).
Can I 'script' this, or do I have to manually contour the work every time I make adjustments to the task ?
Or should I rather split the task into 5 groups of 2 dozers and try to play with dependencies to make it work together ?
2. Can I 'link' Contours ?
In my above example, if I have a pool of 16 Mechanics, can I have a Task "Dozers Maintenance" that will automatically follow the number of dozers working and adapt the number of Mechanics accordingly each day ? (let's say 1 Mechanic for 2 dozers)
3. Can I define Tasks with higher priority (I tried but couldn't make it work), assign them resources and them define other Tasks with lower priorities where Project would automatically assign only the remaining resources to those Tasks ?
Of my 16 Mechanics, some are tied up with dozers, but not all, can I define an other Task that is a lower priority and get Project to automatically assign just the remaining number of Mechanics available ?
Boy, the fun just doesn't stop does it. Let me take a crack at your questions, others may chime in.
1. Can you "script" your custom contour? Yes, you could create a macro that will set up the desired custom contour and then run the macro any time you wan to apply it. However, getting the algorithm set up to do what you want may be challenging. But yes, it can be done.
2. Can you link contours? No, not in the normal linking sense as other links in Project but yes, you could incorporate the dozer maintenance coordination into the custom contour algorithm in the macro.
3. No, Project will not assign resources for you. The resource leveling functionality can move resources to alleviate overallocation but it cannot assign other resources to fill in work gaps. Remember, Project has zero intelligence, it simply schedules tasks given basic inputs from the user. You can "give it some intelligence" with a macro, but that's it.
Hope this helps.
If you think you need to do a few changes and need an accurate profile, I would have 5 ramp up tasks and 5 ramp down tasks plus one main task. All under a summary task. This will be more flexible, but of course the downside is more tasks.
Rod Gill Project Management
I'm self-taught but I highly recommend Rod Gill's book. He provided a link in his response. Rod's book includes many examples of Project VBA code. And if you have specific questions, we are more than happy to assist you. That's why we are here. You can post those questions here but it is preferable if you post to the Project customization and programming forum at, http://social.technet.microsoft.com/Forums/projectserver/en-US/home?forum=project2010custprog.
This seems to be the key in MS Project: Splitting tasks into 'elementary tasks' or 'sub-tasks' until the point where we can pretty much shape things the way we need. This is what some refer to as the 8/80 rule, right ?
In my project, I have some tasks that might last over 60 days (such as "Opening Access" with my dozers), so I am well out of the 8/80 scheme. At first glance, it makes little sense to me to split the task (except as you suggest to better shape the ramp-up and ramp down) but further than that, I am totally unable to predict what will be the exact daily work of each dozer. It so much depends on how many operators are available every day, how many dozers are broken down and under repair, how much the access have deteriorated over night (there might have been a sand storm and some could have to re-work the section they've been opening for the past few days). So I may have a very detailed 'theoretical' plan broken down for each dozer and showing for each one their daily work section, but after an hour of work, this plan will be worthless.
Should I just keep my 2 months task, knowing it is only as accurate as these things can be, or should I try harder and split further ?
Thanks for your help
I know this is a response for Rod and I'm not pretending to answer for him but allow me to throw my two cents in on this part of the discussion.
One of the biggest problems about putting together a schedule is to get the right balance of detail. Too little and the plan is nothing more than an outline. Too much and the plan is either so bogged down in the minutiae or so complex that managing to/with it is hopeless. Attempting to detail plan projects with numerous "unknowns" is challenging. Construction projects and engineering development projects are two prime example of projects with many unknown unknowns. My advice, for whatever it's worth, keep the plan at a workable level and be prepared to re-structure it often.
I understand what you are saying, I'm still struggling to define the right level of detail and this is indeed challenging as I don't want the follow-up and tracking of my project to become a full-day job!
I have no problem re-structuring often but I also need to have consistency so that I can compare my current status to an Initial Plan or Baseline that is actually comparable to. I guess this is the very essence of making an efficient use of MS Project. I am clearly not there yet.
Thanks for pointing out the link, I have ordered my copy of the book and I can't wait to get started to see what I can do with it.
Talking about construction projects, I am in the seismic industry, so I guess this is probably the closest type of project I can relate to. Some projects are just a pile of unknowns and rolling a dice is as good as any guess. I found that many books or blogs that deal with MS Project often relate to IT Development or Software Programming Projects and these examples are way too far from my model. Would you know of any resources that deal more specifically with 'Construction type' Projects such as mine ?
Or to put it another way, could it be that MS Project is just not the software I need ? Are there any other tools that could better suit my needs ? (this might not be the right place to ask this question, though...)
Again, thanks for your time and help, it is really appreciated.
You're welcome and thanks for the feedback. I assume Rod will check in later today.
I'm glad you decided to go ahead and get Rod's book. I believe VBA is one of the most powerful tools available for Project, and other apps for that matter, and I've always been grateful that Microsoft includes it as part of the basic app.
I have to believe there are examples of construction related projects using Project but I can't give you any direct references. The question has been asked previously in the Project forums. You could do a search and see what you can find, or perhaps someone else will jump into the thread and offer some references.
In my opinion, Project is better suited to a project that can be readily defined (i.e. start, do these steps and you're done). When the uncertainties begin to creep in, the planning starts to get complicated. Can Project be used for day-to-day or minute-to-minute activities? I guess so. We had a post a few months back where a user was actually using Project to plan her daily schedule. Wow, that was a bit over-the-top in my view, but a few people actually responded with suggestions. And I've also seen and helped users use Project in very unique ways so it's really up to the user's imagination, maybe with a little help from us.